(17) As a result of the typical sparsity in such dataset, matrix W will also be sparse.
It may not be overlooked that Psb28 is a very short protein containing 112 computable amino acid residues in the dataset matrix, from which only 82 characters are parsimony informative.
For dataset matrix X of size l× m with l probes and m genes we denote the l×1 vector of response value y.
Based on these latent components, PLS predicts response features (the task of regression) and reconstructs an original dataset matrix (the task of data modeling) at the same time.
This filter may be applied afterwards (and/or thus create an additional dataset) matrix size 128 × 128 up to 256 × 256 zoom = 1 2D or 3D OSEM reconstruction with sufficient convergence (generally, product of iterations and subset larger than 50 is sufficient) 5-mm FWHM Gaussian filter in all directions or equivalent filters.
This filter may be applied afterwards (and/or thus create an additional dataset) matrix size 128 × 128 up to 256 × 256 zoom = 1 Philips systems (Gemini, Gemini ToF): There are two options: Gemini TF: LOR-TF-RAMLA ("Blob-OS-TF") with 'normal' smoothing setting.
We used two separate datasets: matrix A (for higher-level analyses) and matrix B (for species-level analyses within Pseuduvaria).
The datasets (matrix and trees resulting from the phylogenetic analyses) supporting the results of this article are available in the TreeBase repository, http://purl.org/phylo/treebase/phylows/study/TB2 S14132.
For the BioHEL classifications on each dataset, confusion matrices (that identify, treatment by treatment, how the samples were predicted) were generated to understand which samples were predicted correctly.
Then, for each simulated dataset, new matrices were derived by multiplying the μ k ⋆ s by a constant c ∈ { 0.1, 0.2, 0.3, …, 1 } to reduce the signal-to-noise ratio.
